From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 02 Nov 2010 13:11:24 +0000
Subject: Re: [BUG] in 2.6.37-rc1-ish in the serial driver
Message-Id: <20101102131124.GA5583@linux-sh.org>
List-Id: <linux-sh.vger.kernel.org>
References: <Pine.LNX.4.64.1011021002020.4552@axis700.grange>
In-Reply-To: <Pine.LNX.4.64.1011021002020.4552@axis700.grange>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: linux-sh@vger.kernel.org

On Tue, Nov 02, 2010 at 10:15:30AM +0100, Guennadi Liakhovetski wrote:
> Got the following Oops:
> 
> [  203.203125] Backtrace:
> [  203.203125] [<c002b0b8>] (__bug+0x0/0x30) from [<c008b5bc>] (cache_alloc_refill+0x128/0x4ec)
> [  203.203125] [<c008b494>] (cache_alloc_refill+0x0/0x4ec) from [<c008ba10>] (__kmalloc+0x90/0xd8)
> [  203.203125] [<c008b980>] (__kmalloc+0x0/0xd8) from [<c014bc14>] (tty_buffer_request_room+0xbc/0x120)
> [  203.203125]  r7:00000100 r6:00000001 r5:00000000 r4:ce7e2400
> [  203.203125] [<c014bb58>] (tty_buffer_request_room+0x0/0x120) from [<c015d8f0>] (sci_rx_interrupt+0x12c/0x3c8)
> [  203.203125] [<c015d7c4>] (sci_rx_interrupt+0x0/0x3c8) from [<c015df4c>] (sci_mpxed_interrupt+0x9c/0x104)
> [  203.203125] [<c015deb0>] (sci_mpxed_interrupt+0x0/0x104) from [<c0061dec>] (handle_IRQ_event+0x2c/0xe8)
> [  203.203125]  r7:00000050 r6:00000000 r5:00000000 r4:ce7e3ae0
> [  203.203125] [<c0061dc0>] (handle_IRQ_event+0x0/0xe8) from [<c0063f14>] (handle_level_irq+0xac/0x11c)
> [  203.203125]  r7:c02e9b48 r6:c002069c r5:00000050 r4:c02f9790
> [  203.203125] [<c0063e68>] (handle_level_irq+0x0/0x11c) from [<c0027078>] (asm_do_IRQ+0x78/0x9c)
> [  203.203125]  r5:00000000 r4:00000050
> [  203.203125] [<c0027000>] (asm_do_IRQ+0x0/0x9c) from [<c0027b74>] (__irq_svc+0x34/0x80)
> [  203.203125] Exception stack(0xc02e7f50 to 0xc02e7f98)
> 
> The kernel is 2.6.37-rc1 based, with a few video patches. Unless I'm 
> wildly corrupting memory somewhere, they shouldn't be related. Although, I 
> am slightly concerned by "LR is at release_console_sem+0x188/0x19c", even 
> though release_console_sem() doesn't reoccur in the backtrace. This 
> _might_ indicate a problem with the framebuffer code.
> 
This certainly looks like memory corruption, can you try with slab
debugging enabled? It would also help to see your .config.